This study will investigate and compare the epidemiology and etiology of substance use and abuse in Washington State and the State of Victoria in Australia. Data will be collected from statewide probability samples of 5th, 7th, and 9th grade students. The 1000 students at each grade level in each state will constitute a longitudinal panel to be surveyed three times at annual intervals. The study will utilize a recently developed instrument capable of cost-efficient measurement of a comprehensive range of risk and protective factors, substance use, and other problem behaviors This instrument is appropriate for a broad age range of youth (age 10-20) providing unique information on the level and comparative strength of these predictors at different developmental stages. The comparison of these processes in Australia and the U.S. will allow for the investigation of the cross national validity of these predictors. The international comparison will also investigate whether cultural differences between the two countries predict substance use and abuse, perhaps leading to the discovery of new risk factors for substance abuse. This study has the following aims: To examine the cross-national validity of survey measures of a comprehensive range of risk and protective factors and substance use and other problem behaviors; To compare the levels of risk and protective factors and prevalence of substance use (substance initiation, frequency of use and heavy use) and related problem behaviors in the two states; To examine the similarities and differences in the pathways to initiation, progression and maintenance of drug using behavior in the two states; To examine the extent and effect of cultural differences between the two states on levels of risk and protection and patterns of initiation, progression, and maintenance of substance use; Specifically, to examine the extent and effect of differences in normative messages of harm reduction versus abstinence in prevention programs and school policies; To test the comparative fit of the Social Development Model (Catalano and Hawkins, 1996) in two states.